Azure AI Document Intelligence
An Azure service that turns documents into usable data. Previously known as Azure Form Recognizer.
2,107 questions
How do I specify key-value pair detection when using document intelligence prebuilt-layout model (Java SDK)?
![](data:image/svg+xml, %3Csvg xmlns='http://www.w3.org/2000/svg' height='64' class='font-weight-bold' style='font: 600 30.11764705882353px "SegoeUI", Arial' width='64'%3E%3Ccircle fill='hsl(73.60000000000001, 45%, 17%)' cx='32' cy='32' r='32' /%3E%3Ctext x='50%25' y='55%25' dominant-baseline='middle' text-anchor='middle' fill='%23FFF' %3EN%3C/text%3E%3C/svg%3E)
Nauhar
0
Reputation points
The Document Intelligence general document model is now deprecated with the 2023-10-31 preview, and the documentation specifies that I should use the Layout model with the optional query string parameter features=keyValuePairs enabled if I want to extract key-value pairs, selection marks, text, tables, and structure from documents. I'm using the Java SDK, and I am trying to do the following :
SyncPoller < OperationResult, AnalyzeResult > analyzeDocumentPoller = client.beginAnalyzeDocumentFromUrl("prebuilt-layout", "insert-document-url-here");
where can I specify features=keyValuePairs so that the model can detect the key value pairs of the document as well? the above code does not detect the key value pairs. I tried doing something like this, but I get an error saying the requested model was not found:
SyncPoller < OperationResult, AnalyzeResult > analyzeDocumentPoller = client.beginAnalyzeDocumentFromUrl("prebuilt-layout?features=keyValuePairs", "insert-document-url-here");
Azure AI Document Intelligence
{count} votes
-
romungi-MSFT 48,911 Reputation points Microsoft Employee Moderator2023-12-26T06:23:19.6533333+00:00 @Nauhar I think you can refer this sample from the SDK repo to pass the DocumentAnalysisFeature.KEY_VALUE_PAIRS and the same in response. The reference for available features is available here.
SyncPoller<AnalyzeResultOperation, AnalyzeResultOperation> analyzeLayoutResultPoller = client.beginAnalyzeDocument("prebuilt-layout", null, null, null, Arrays.asList(DocumentAnalysisFeature.KEY_VALUE_PAIRS), null, null, new AnalyzeDocumentRequest().setBase64Source(Files.readAllBytes(invoiceDocument.toPath()))); AnalyzeResult analyzeLayoutResult = analyzeLayoutResultPoller.getFinalResult().getAnalyzeResult(); analyzeLayoutResult.getKeyValuePairs().forEach( keyValuePair -> { DocumentKeyValueElement key = keyValuePair.getKey(); DocumentKeyValueElement value = keyValuePair.getValue(); if (key != null) { System.out.println("- Key: " + key.getContent()); List<BoundingRegion> boundingRegions = key.getBoundingRegions(); boundingRegions.forEach(boundingRegion -> { System.out.printf(" Bounding regions page: %s, polygon: %s%n", boundingRegion.getPageNumber(), boundingRegion.getPolygon()); }); }The GH repo samples folder has examples for other scenarios for reference. I hope this helps!! Thanks!!
-
romungi-MSFT 48,911 Reputation points Microsoft Employee Moderator
2023-12-27T05:30:29.2166667+00:00 @Nauhar Did the above comment help to answer your question? Thanks!!
-
Nauhar 0 Reputation points
2023-12-31T04:33:06.0533333+00:00 @romungi-MSFT thank you for your response. Unfortunately, your solution did not work for me. When I copy the following code into my IDE, I get an error saying "Cannot resolve method 'beginAnalyzeDocument(String, null, null, null, List<T>, null, null, AnalyzeDocumentRequest)'"
Was the code you sent me from a previous version of the document intelligence sdk? I'm currently using version 4.1.4 of the sdk.
SyncPoller<AnalyzeResultOperation, AnalyzeResultOperation> analyzeLayoutResultPoller = client.beginAnalyzeDocument("prebuilt-layout", null, null, null, Arrays.asList(DocumentAnalysisFeature.KEY_VALUE_PAIRS), null, null, new AnalyzeDocumentRequest().setBase64Source(Files.readAllBytes(invoiceDocument.toPath())));
Sign in to comment
Sign in to answer